The Bug at Volkswagen

Corporate governance scholarship has long consid­ered the problems that arise in public companies with dispersed ownership. But the automaker Volkswagen does not suffer from a dispersed ownership structure. In fact, it has several strong and highly active owners. The Porsche and Piëch families have been involved with the company for many years and own 31.5% of Volkswagen’s equity. The German state where the company is headquartered, Lower Saxony, holds 12.4%, and an outside investor, Qatar Holding, owns 15.4%.  With such powerful economic incentives in not one but three actors, management should have been subject to the kind of exacting oversight that could readily ferret out misconduct.

Yet problems arising from the separation of ownership and control are not the only issues that can lead to corporate wrongdoing—and VW is a striking case-in-point. It has recently been revealed that, despite the influence and financial stakes of its three largest investors, the company has long been involved in a massive emissions regulation evasion scheme. From 2007 to 2015, the company cheated on federal emissions tests through the installation of a so-called “defeat device” on at least eleven million of its vehicles.  With the device in place, the company fooled regulators and hid emissions of as much as 40 times the allowable amount of nitrogen oxide.

Given the company’s concentrated ownership structure, the problem cannot be attributed to the separa­tion of ownership from control.  In the discussion of the VW emissions scandal that follows, which we elaborated further in a recent article in the Journal of Applied Finance, we explore the causes of the scandal, highlighting issues that arise outside of the conventional dispersed ownership paradigm.

Board Structure

Under the German Codetermination Act (Mitbestimmungesetz), all companies with 2000 or more employees must have both shareholder and employee representatives. The Volkswagen supervisory board has long met this requirement. Its board has 20 members, with ten elected by the shareholders and the remaining ten selected directly by the workforce. This board is mainly responsible for selecting and monitoring the separate “management board” that in turn runs the company.

The inclusion of labor representatives on the board presents serious difficulties for the employee-monitoring function of management. The board and management are responsible for supervising employees. Ordinarily, it is management’s duty to set compensation levels, prevent and penalize shirking and misfeasance, and efficiently coordinate labor. But with labor representatives on the board, the party that is susceptible to shirking is also the party responsible for the preventive monitoring. As a result, oversight is likely to be less efficient. Although the primary aim of monitor­ing is meant to identify and constrain shirking, it also has the effect of providing oversight designed to ferret out illegal acts. By so doing, effective oversight reduces the likelihood of both misfeasance and malfeasance. But the conflict of inter­ests and incentives inherent in the German model may have undermined the oversight function, and reduced attention to appropriate legal compliance.

Shareholders representatives in co-determined firms often manage much of the board work informally or directly with management in ad hoc committees. But such a practice has the drawback that the board’s access to essential information about these projects and its ability to make reasoned decisions about them are often severely limited. Practices such as giving supervisory directors board materials only slightly before the meeting, reducing the number of meetings of the board and its formal committees, or increasing the board’s size to unwieldy numbers that encourage free-riding and impede efficient meeting management are common in German co-determined companies. As a result of this response, the supervisory board is significantly weakened and its ability to function as a monitoring body is limited.

In effect, then, the German co-determined firm appears to run the risk of becoming something of a headless state, its supervisory board neutered in an effort to avoid inefficiencies in the manner in which it structures the decision-making process. What is left is just a balancing of powers, with policy largely determined in the confluence of each interest. Of course, the massive regulatory scandal that occurred at Volkswagen was in no party’s interest. But it occurred under all of their watch because, without a strong supervisory board, nobody was looking after the common enterprise.

At VW, the co-determination structure interacted with the controlling and government shareholdings. The Porsche and Piëch families were able to maintain significant influence over the company through their control of 50.7% of the shareholder vote. Thanks to this control, they were and continue to be able to choose the chairperson of the supervisory board. Under his leadership, the company appears to have followed, without question, his “desire to make VW the best and biggest carmaker in the world.”  This objective happened to be aligned with the goals of both labor and the other major constituency, government.

The Controlling Shareholders

Volkswagen has a controlling shareholder that uses both pyramidal and dual-class structures to enhance control while insulating itself from the economic consequences of company transactions. Through Porsche Automobil Holding SE, an investment holding company, members of the Porsche and Piëch families own shares that represent 31.5% of the equity interest in VW. This interest is leveraged through a dual-class structure. Thanks to its ownership of preferred superior voting rights securities, the Porsche and Piëch families control 50.7% of the voting rights. This control is further leveraged through the use of a pyramid structure. Half of the shares of Porsche Automobil Holding are ordinary voting shares, and the other half are non-voting shares. The Porsche and Piëch families own the ordinary shares, but not the non-voting shares. Through these structural devices, the families control five board seats (half of the investor representatives on the super­visory board), while limiting their own economic exposures.

The nature of the agency problem that affects controlled companies differs from the agency problem that tends to reduce the value of, and is the main corporate governance challenge facing widely held companies in the United States. For the outside investors in U.S. public companies, the blunt­ness of the market for control as a disciplinary mechanism and the difficulties inherent in disaggregated shareholders’ use of formal power through proxy voting create an inabil­ity to effectively monitor the officers and directors of such companies. In this sense, the directors and officers of U.S. companies have de facto control, and the agency concern in such cases is thus about management opportunism.  This is sometimes called a vertical agency conflict.

In controlled companies, by contrast, the controlling shareholder has an economic incentive to monitor manage­ment because it stands to capture a significant proportion of any increase in share value that results from such monitor­ing. For this reason, a controlling shareholder has at least the potential to be a more effective monitor of corporate management than is possible with the mechanisms available in the widely held model.

So, in a controlled company, then, management oppor­tunism is, at least in theory, much less likely to be a significant problem. But there is another kind of agency problem that can arise in such companies—namely, oppor­tunism by the controlling shareholder at the expense of the minority shareholders. The incentive for such opportunistic behavior arises from the fact that not all of the costs borne by the company are “internalized” by the controller because the controller does not own 100% of the equity. This creates the potential for non-proportional diversions of pecuniary benefits from the company to the controller through self-dealing transactions.  The problem that arises here is sometimes referred to as a horizontal agency conflict.

The compliance problems at VW do not appear to stem from the diversion of value through self-dealing transactions designed to enrich the controlling shareholders. The real issue that led to the scandal appears to have originated in the ways that nonpecuniary private benefits influenced the management and strategy of the company. One member of the controlling families, Ferdinand Piëch, was able to take leadership of the company and direct it in pursuit of his own ambitions for industrial domination. It was in the course of that pursuit that the emissions scandal occurred. Agency problems of this non-pecuniary type can be especially pernicious. To be sure, the top managers of widely held public companies may have a similar predilection for activities such as empire building, but there is at least some constraint on inefficient behavior from markets and the shareholder franchise. In the case of controllers, though, that constraint is entirely lacking. Controllers are immune from proxy contests and can “just say no” to any takeover offer. They are, in large part, free to set the company’s strategy largely as they wish, subject only to a different, much less demanding set of constraints, that we discuss later.

Piëch, the grandson of Ferdinand Porsche, the founder of the Porsche company, owns 10% of Porsche Automobil Holding, the family investment fund. From 1993 until 2002, he was the CEO of VW. After he retired from that position, he became chairman of the supervisory board. He was chairman when the emissions scandal occurred, stepping down only earlier this year after a failed attempt to oust the then-CEO. At some point, Piëch,’s interest in VW appears to have become largely non-pecuniary, his goal was no longer (if it ever was) to create more wealth for this family (or his shareholders). A number of commentators have claimed that, for him, “what is important is power, not money”—and that his real underlying objective was to create the world’s largest automobile manufacturer.  In this quest to expand his company’s size and market share, Piëch assembled a 12 brand automotive empire—one that included the likes of Audi and Bugatti as well as VW—that aimed to displace Toyota as the world’s largest.

As Piëch built his empire, seemingly without much regard for profitability and shareholder value, the means by which the company achieved its growth may have been irrelevant. As the controlling share­holder, with the cooperation of the labor unions and the government—which both sought greater employment— Piëch had for years the effective power to direct all corporate activities toward this goal. The emissions actions at VW, unfortunately, are consistent with this paradigm of growth at all costs, including as it turns out, outright misfeasance. And in the face of such pressure for growth, the minority-shareholder interests on VW’s board failed to provide the appropriate counter-weight to the dominance of its chairman.

The Government as a Shareholder and Representative

In addition to the controlling shareholder, a major structural problem with the VW board was the presence of the government as a shareholder. When a dominant shareholder happens to be a government, the goals of that shareholder are unlikely to match those of the remain­ing public equity investors. To the shareholders, the concept of long-term returns on capital is paramount, but govern­ments are typically motivated by political considerations, particularly the employment of its citizens. The government of Lower Saxony has had significant influence over the affairs of VW and its board since the company’s creation. Volkswa­gen was originally a state-owned enterprise until, in 1960, the “Volkswagen Law” was enacted by the German government that enabled the company to be privatized while allowing Lower Saxony to retain a 20% voting interest in the company (despite the fact that their actual equity position was only 12.4%) in order to maintain government influence. The VW Law also required the vote of 80% of the shareholders to approve any corporate action by the company and mandated the appointment of two governmental board members. In this way, Lower Saxony, despite the divestment, retained control over the enterprise.

The princi­pal interest of political leadership is the retention of power through reelection, and governments tend to be reelected when there is popular content created by high employment. Therefore, the main motive of the governmental shareholder is to maximize employment, if necessary at the expense of shareholders. And so in cases where the board is composed in part of political representatives, even very limited government ownership can create a skewed board decision-making process. Furthermore, the incentives of governmental directors to expand the employee base at the expense of the proper corporate objective—that is, long-run profitability and value maximization—also has the potential to divert the focus of other directors from value maximization out of the fear that the government will campaign for their replacement should they fail to comply with its goals. With government represen­tatives on the board, the other board members are likely to feel pressure to accommodate them by making decisions under the constraint of the government’s incentives, particularly the preservation of jobs. In such a case, the government effectively acts as a controlling shareholder and, when combined with the family-controlled votes, the potential for conflicts with outside investors is compounded. Either of these blockholders can veto any major decisions. This extraordinary mix of incentives seems quite capable of giving rise to and then tolerating a culture in which there was an incentive to cut corners and falsify emission rates, in order to satisfy both the government’s expectations for jobs and Piëch’s goal of market dominance.

Although there is no evidence that German politicians had any involvement in the emissions scandal at this time, critics claimed that “Berlin fought hard to shield its carmakers from closer scrutiny and, in a high profile clash with its European partners two years ago, from emissions targets.”  Even in the wake of the emissions scandal, German Prime Minister Merkel defended the stance as necessary to protect employment in the sector—a clear indication of the government’s employment-based goal in its investment in VW, the region’s single largest employer.  It seems clear that neither the government nor the controller, in their single-minded pursuit of higher employment, made compliance oversight a primary goal, which left the company vulnerable to the lapse whose consequences it now confronts.

Power Sharing and Common Purpose

A power struggle earlier this year illus­trated in dramatic fashion that control cannot be exercised unilaterally within the VW governance structure. Piëch, disappointed with the company’s performance under Winterkorn, the CEO that he had hand-selected several years earlier, announced that he was dissatisfied and attempted to oust him. But, it was immediately apparent that, acting alone, he did not have the votes to remove Winterkorn. Even within the control­ling families, the members do not vote as a single block nor do they always agree. Some family members, such as his cousin Wolfgang Porsche, did not support the change. Additionally, Stephen Weil, the premier of Lower Saxony, pledged his support, and the State’s 20% vote, for Winter­korn. Then, the leader of the works council pledged its support for Winterkorn as well. Therefore, Piëch’s attempt to replace the CEO backfired and Winterkorn survived. And Piëch was forced to resign from the supervisory board.

The co-determination structure compels this form of power-sharing. Before he was ousted, Piëch’s interest in growing VW into the world’s “biggest carmaker” appeared to be largely if not completely consistent with the interests of the board representatives of labor and govern­ment. Most important, the policy favoring growth without regard for profitability promised to expand Volkswagen’s employee base. And thus this failed attempt to unseat Winter­korn was at bottom reflective only of a new alignment of the employees and works council with management. Before the scandal and after the departure of Piëch, Winterkorn had the workers’ support: “Together with him we have written an unprecedented success story,” works council chair Osterlich said. This support was absolutely necessary for a CEO like Winterkorn to keep his job. The demonstration of the power of the works council and Lower Saxony, through their ability to influence the VW leadership, has made clear the reality that Winterkorn and the supervisory board are dependent on the continued appeasement of both groups to implement their objectives.

Piëch, who led the company during the period that it was misleading regulators, was long an ideal company leader from labor’s perspective. His primary focus was not on making money. Rather, he set out to create an empire through aggressive brand acquisitions. Piëch believed that “It is not possible to bring a company to the summit while maintaining harmony,” so he had to make a choice between labor and shareholders. He chose labor.

The unlikely alliance between an industrial tycoon and labor was possible only within the unique structure of VW and the German law of co-determination. The ability of shareholders to control the direction of the company—for example, through the market for corporate control—was severely curtailed by the voting structures on the board. Piëch’s own economic incentives were sharply diminished by the separation of control rights from cash-flow rights. He owned VW through both a pyramid structure and dual-class shareholdings. For him, the private benefits of empire-building clearly outweighed the sacrifice of potential wealth, most of which would have been shared with outsiders. This powerlessness of shareholders to influence the direction of the company, along with Piëch’s interest in building the world’s largest automaker, coincided with labor’s interests and gave opportunity and reason for the alliance.

Conclusion

The true puzzle of the emissions scandal is that highly interested parties permitted a compli­ance failure that not only benefited none of them, but ended up imposing very large costs of all of them, including labor and local government. Regulatory non-compliance did not in the end further the objectives of any party. However, because the co-determination structure forced an alliance between the shareholder-controller, management, and labor—one that was reinforced by the participation of local government—there was no one to monitor any of those parties. Compliance systems require a relationship that includes healthy distrust and skepticism between the board and management, and management and labor—a relation­ship that is less likely to develop when the parties’ primary objectives are all so closely aligned.

The problems at VW resulted from the conflict of objectives that is inherent in the structure of the VW board. Policy was determined by the individual interests of each party that was represented on the board. Thus large-scale consideration of the company’s ethical posture seems to have been secondary. This resulting blindness to the potential for corporate malfeasance on what turned out to be a global scale is, in our view, the result of the confluence of three problems that can be traced to the composition of VW’s board. The presence of a dual-class controlling share­holder, the government as a major equity-holder and putative board member, and, finally, the German corporate model of co-determination all contributed to this massive board fail­ure in oversight. The scandal gives investors and directors much to reflect on. Ultimately, it is a testament to the importance of board composition, theory, and structure in helping to build and maintain a corporate culture that promotes integrity and, in the final analysis, the long-run success of the company.

The preceding post comes to us from Charles M. Elson is Edgar S. Woolard, Jr., Chair in Corporate Governance and Director, The John L. Weinberg Center for Corporate Governance, University of Delaware, Craig K. Ferrere, JD candidate at Harvard Law School, and Nicholas J. Goossen an Undergraduate and Research Assistant at the John L. Weinberg Center for Corporate Governance, University of Delaware.  The post is based on their paper, which is entitled “The Bug at Volkswagen: Lessons in Co-Determination, Ownership, and Board Structure” and available here.